Using the uncertainty handling CMA-ES for finding robust optima

Algorithms that search for robust optima often evaluate the effective fitness (robust fitness) based on stochastic approximation schemes. In this setting, finding robust optima can be recast as an optimization problem with an/a uncertain/noisy objective function. This paper studies whether state-of-the-art uncertainty handling techniques, proposed in the context of optimizing noisy objective functions, can be applied for finding robust optima. In this paper, the UH-CMA-ES is modified to handle approximations of the effective fitness. This modified approach, named RO-UH-CMA-ES, is evaluated empirically and compared to other schemes that aim to find robust optima. The experiments on multiple benchmark problems show that the RO-UH-CMA-ES yields comparable results for multi-modal problems and it outperforms other schemes on unimodal problems.

[1]  Shigeyoshi Tsutsui,et al.  Genetic algorithms with a robust solution searching scheme , 1997, IEEE Trans. Evol. Comput..

[2]  Christian Igel,et al.  Hoeffding and Bernstein races for selecting policies in evolutionary direct policy search , 2009, ICML '09.

[3]  Thomas Bäck,et al.  A robust optimization approach using Kriging metamodels for robustness approximation in the CMA-ES , 2010, IEEE Congress on Evolutionary Computation.

[4]  Ofer M. Shir,et al.  Adaptive Niche Radii and Niche Shapes Approaches for Niching with the CMA-ES , 2010, Evolutionary Computation.

[5]  Kai-Yew Lum,et al.  Max-min surrogate-assisted evolutionary algorithm for robust design , 2006, IEEE Transactions on Evolutionary Computation.

[6]  Shigeyoshi Tsutsui,et al.  A comparative study on the effects of adding perturbations to phenotypic parameters in genetic algorithms with a robust solution searching scheme , 1999, IEEE SMC'99 Conference Proceedings. 1999 IEEE International Conference on Systems, Man, and Cybernetics (Cat. No.99CH37028).

[7]  Benjamin W. Wah,et al.  Scheduling of Genetic Algorithms in a Noisy Environment , 1994, Evolutionary Computation.

[8]  Acknowledgments , 2006, Molecular and Cellular Endocrinology.

[9]  S. Tsutsui,et al.  Effects of adding perturbations to phenotypic parameters in genetic algorithms for searching robust solutions , 2003 .

[10]  Jürgen Branke,et al.  Efficient search for robust solutions by means of evolutionary algorithms and fitness approximation , 2006, IEEE Transactions on Evolutionary Computation.

[11]  Jürgen Branke,et al.  Creating Robust Solutions by Means of Evolutionary Algorithms , 1998, PPSN.

[12]  Nikolaus Hansen,et al.  Completely Derandomized Self-Adaptation in Evolution Strategies , 2001, Evolutionary Computation.

[13]  Bernhard Sendhoff,et al.  Evolution Strategies for Robust Optimization , 2006, 2006 IEEE International Conference on Evolutionary Computation.

[14]  Thomas Bäck,et al.  Evolution strategies applied to perturbed objective functions , 1994, Proceedings of the First IEEE Conference on Evolutionary Computation. IEEE World Congress on Computational Intelligence.

[15]  Thomas Bäck,et al.  On the limitations of adaptive resampling in using the student's t-test evolution strategies , 2009, GECCO '09.

[16]  Thomas Bäck,et al.  An Archive Maintenance Scheme for Finding Robust Solutions , 2010, PPSN.

[17]  J. Branke.  Reducing the sampling variance when searching for robust solutions , 2001 .

[18]  Petros Koumoutsakos,et al.  A Method for Handling Uncertainty in Evolutionary Optimization With an Application to Feedback Control of Combustion , 2009, IEEE Transactions on Evolutionary Computation.

[19]  Thomas Bäck,et al.  Robust design of multilayer optical coatings by means of evolutionary algorithms , 1998, IEEE Trans. Evol. Comput..

[20]  Peter Stagge,et al.  Averaging Efficiently in the Presence of Noise , 1998, PPSN.

[21]  Thomas Bäck,et al.  Exploiting Overlap When Searching for Robust Optima , 2010, PPSN.

[22]  Christian Igel,et al.  Uncertainty handling CMA-ES for reinforcement learning , 2009, GECCO.

[23]  Jürgen Branke,et al.  Evolutionary optimization in uncertain environments-a survey , 2005, IEEE Transactions on Evolutionary Computation.

[24]  H. Beyer.  Evolutionary algorithms in noisy environments : theoretical issues and guidelines for practice , 2000 .

[25]  Bernhard Sendhoff,et al.  Functions with noise-induced multimodality: a test for evolutionary robust Optimization-properties and performance analysis , 2006, IEEE Transactions on Evolutionary Computation.

[26]  Bernhard Sendhoff,et al.  Robust Optimization - A Comprehensive Survey , 2007 .

[27]  Hans-Georg Beyer,et al.  Why noise may be good: additive noise on the sharp ridge , 2008, GECCO '08.